Blood flow is known to increase the lateral motions of platelets and to be associated with a decrease in near-wall concentrations of red cells in a region commonly referred to as the marginal layer. (Here this term implies the entire near-wall rgion where there is a significant change of red cell concentration (local hematocrit) and not just the peripheral plasma layer.) Recent studies by us and others show that the marginal layer is populated by platelets at a concentration significantly in excess of that in the main portion of the flow. Factors that control this local platelet concentration are relatively unknown and the rates at which the region is established or can be replenished are unstudied. Given the key role of the platelet in the coagulation and thrombotic sequences, an understanding of the rheological factors that control this near-wall platelet excess is needed to advance the total understanding of the thrombotic process. Proposed work will study the near-wall excess by capturing the distribution and orientation of red cells and platelets (or platelet-sized latex particles) in a capillary tube. The cryogenic freezing technique proven by Phibbs and coworkers will be used for this purpose. Experiments will use a bolus flow since the front of the bolus facilitates study of the initial events of the near-wall excess and, when the bolus is long enough for the wall region to be developed, it allows easy measurements of the fully developed near-wall region. Further, recirculation of the near-wall platelet excess at the rear of the bolus provides an opportunity to study the spread of platelets without inducing a gradient by reactive means. Videomicrography and a microcomputer-based digitizing apparatus will provide a labor-efficient means of determining profiles of platelet concentration from sections of the frozen tube. Using these profiles, various models of platelet transport will be examined. The most ambitions models are expansions of the simple random walk models currently used for modelling the flow-enhanced lateral platelet motions. These expansions look to modelling the difference of red cell and platelet behavior in the marginal layer and the different likelihood of collision at different positions in the marginal layer. Also, the models will examine simpler concepts such as whether basing the concentration of platelets on the available plasma at each location improves the convective diffusion theory for platelet transport.